How much are iceberg orders a detriment on tick charts?

[CyJackX]

I tend towards volume charts, but I recently opened up a tick chart just to see what else it offered. One, I was very surprised by how many trades actually occur in a given time period, it was much smaller than I thought. I think I was overestimating the number of players in the market at any moment.

But two, I was thinking about iceberg orders, which I imagine are pretty much the way anybody trying to move a lot of volume executes. Someone trying to move 1000 contracts 10 at a time will generate 100 trades at least, which will show up completely different on a volume chart.

Are most market-moving orders iceberg orders? Can tick charts still be believed based on that?

 

[Tim Smith]

IMHO, tick charts are just noise and best left alone.

The days of watching volume charts are also somewhat over with the advent of technology and algorithms, i.e. what you're seeing on the volume charts might not be what you think it is. Lots of the stuff written in books about volume analysis is kind of obsolete in today's markets.

For example, you say "I think I was overestimating the number of players in the market", well how about considering the fact you weren't, but the way they are represented in what you are seeing is ? i.e. at one end you've got the algorithms jumping around "Flash Boys" style, but at the other end you've got the retail traders/investors.

At the retail end, you can get internalization and all sorts of things going on that skew your interpretation of events if you're trying to invent a story out of what you're seeing on the short-term volumes.

Similarly, you say "Someone trying to move 1000 contracts 10 at a time will generate 100 trades at least".

That may well be. But who's saying they need to move them all "right now" ? If you're a pension fund, unit trust, or just a big retail trader, you might decide to sell/buy a holding but you'll probably be very happy to let the dealing desk hang onto your order and flush it slowly into the market over the course of the day (or maybe over a few days).

People wanting to shift large volumes are generally not stupid, they have systems and processes in place to make sure you don't find out about it until they're done (assuming you manage to find their trades amongst the noise in the first place).

Most of the noise you see on tick level data is probably all those algo's scalping the cents off other people's trades.

 

[i960]

The days of watching volume charts are also somewhat over with the advent of technology and algorithms, i.e. what you're seeing on the volume charts might not be what you think it is. Lots of the stuff written in books about volume analysis is kind of obsolete in today's markets.

He's talking about CVB charts, not a literal chart of volume.

 

[CyJackX]

IMHO, tick charts are just noise and best left alone.

The days of watching volume charts are also somewhat over with the advent of technology and algorithms, i.e. what you're seeing on the volume charts might not be what you think it is. Lots of the stuff written in books about volume analysis is kind of obsolete in today's markets.

For example, you say "I think I was overestimating the number of players in the market", well how about considering the fact you weren't, but the way they are represented in what you are seeing is ? i.e. at one end you've got the algorithms jumping around "Flash Boys" style, but at the other end you've got the retail traders/investors.

At the retail end, you can get internalization and all sorts of things going on that skew your interpretation of events if you're trying to invent a story out of what you're seeing on the short-term volumes.

Similarly, you say "Someone trying to move 1000 contracts 10 at a time will generate 100 trades at least".

That may well be. But who's saying they need to move them all "right now" ? If you're a pension fund, unit trust, or just a big retail trader, you might decide to sell/buy a holding but you'll probably be very happy to let the dealing desk hang onto your order and flush it slowly into the market over the course of the day (or maybe over a few days).

People wanting to shift large volumes are generally not stupid, they have systems and processes in place to make sure you don't find out about it until they're done (assuming you manage to find their trades amongst the noise in the first place).

Most of the noise you see on tick level data is probably all those algo's scalping the cents off other people's trades.

How does modern tech invalidate volume charting? And yes, I meant constant volume bars like i960 mentioned.

Not sure about your comment re: players in the market. I mentioned seeing far fewer than I thought. Fast or slow, the tick data should indicate that a trade occurred, no? You couldn't be saying that HFTs or somebody gets to completely hide their trade from time & sales? I was seeing hundreds of trades per minute; I was expecting with HFTs et al to be seeing thousands.

They don't necessarily want to move 1000 contracts right now; all I'm saying is that a single trade of 1000 contracts vs 1000 trades of 1 contracts shows up very differently on a tick chart, but looks identical on a CVB chart, and I'm not sure what advantage a tick chart has over CVB.

 

[i960]

They don't necessarily want to move 1000 contracts right now; all I'm saying is that a single trade of 1000 contracts vs 1000 trades of 1 contracts shows up very differently on a tick chart, but looks identical on a CVB chart, and I'm not sure what advantage a tick chart has over CVB.

I guess in this case a tick chart has the advantage of showing you potential hidden liquidity being filled - although that's predicated on individual trades truly representing exactly what it looks like, and in the past that's been an issue: http://emini-watch.com/cme-changes-tick-data-again-mdp-3-0/6728/

 

[fordewind]

IMHO, tick charts are just noise and best left alone.

The days of watching volume charts are also somewhat over with the advent of technology and algorithms, i.e. what you're seeing on the volume charts might not be what you think it is. Lots of the stuff written in books about volume analysis is kind of obsolete in today's markets.

For example, you say "I think I was overestimating the number of players in the market", well how about considering the fact you weren't, but the way they are represented in what you are seeing is ? i.e. at one end you've got the algorithms jumping around "Flash Boys" style, but at the other end you've got the retail traders/investors.

At the retail end, you can get internalization and all sorts of things going on that skew your interpretation of events if you're trying to invent a story out of what you're seeing on the short-term volumes.

Similarly, you say "Someone trying to move 1000 contracts 10 at a time will generate 100 trades at least".

That may well be. But who's saying they need to move them all "right now" ? If you're a pension fund, unit trust, or just a big retail trader, you might decide to sell/buy a holding but you'll probably be very happy to let the dealing desk hang onto your order and flush it slowly into the market over the course of the day (or maybe over a few days).

People wanting to shift large volumes are generally not stupid, they have systems and processes in place to make sure you don't find out about it until they're done (assuming you manage to find their trades amongst the noise in the first place).

Most of the noise you see on tick level data is probably all those algo's scalping the cents off other people's trades.

lol..Volume analysis was dead long time ago.